Forms of Address in Chilean Spanish Kelley Bishop

Jim Michnowicz North Carolina State University, USA

Abstract: The present investigation examines possible social and linguistic factors that influence forms of address used in Chilean Spanish with various interlocutors. A characteristic of the Spanish of Chile is the use of a variety of forms of address for the second person singular, tú, vos, and usted, with corresponding spoken conjugations (Lipski 1994). Previous studies have posited that whereas the pronominal voseo is still stigmatized by Chile’s middle and upper classes, the mixed verbal voseo is gaining ground among the educated sectors of the population (Torrejón 1986, 1991). The current study uses linguistic survey results taken from eighty-one residents of Santiago, Chile, to test these predictions quantitatively. The results show a V-shaped distribution of verbal voseo in terms of age and social class. Young professional-class speakers are those who report using verbal voseo most frequently. Survey data are compared to recorded and observed interactions among speakers. Interactional data suggest that the frequency of vos is under- reported in the survey but that the essential pattern reported by speakers is accurate.

Keywords: Chilean Spanish, forms of address, voseo

1. Introduction mong the features that distinguish many dialects of contemporary Latin American Spanish from those of the Iberian Peninsula is the existence of voseo, an informal Aform of address used with a single interlocutor. The form exhibits variation in terms of acceptability and contexts of use depending on dialect, ranging from being the preferred form in areas such as the River Plate region to being heavily stigmatized, as is the case in areas of Venezuela (Lipski 1994: 159–61). In Chilean Spanish in general, although overt (pronominal) voseo is stigmatized, some studies have suggested an increasing acceptance of verbal voseo among urban speakers (Torrejón 1986, among others). The goal of the present study is to test quantitatively the observations made by previous researchers on Chilean verbal voseo by ap- plying multivariate statistical analysis to data obtained through linguistic surveys and recorded observations. Statistical analysis of this kind provides a more concrete picture of variation in forms of address than that provided by previous work. Additionally, a comparison of survey and observed results will address questions of reliability for different data collection techniques.

2. Origins of Chilean Voseo The use of voseo is one of the most characteristic features of many dialects of Latin American Spanish and has even been called a symbol of linguistic identity for the area (Cisneros E. 1998: 90). In his 1981 study, Páez Urdaneta estimated that approximately 47% of the Latin American population was in contact with voseo and that, geographically, the area where voseo was used was twice that of the area where voseo was not used. As voseo fell out of favor with Peninsular speakers by the eighteenth century, it also fell into disuse among speakers in areas of Latin Bishop & Michnowicz / Forms of Address in Chilean Spanish 415

America in constant contact with Spain (Mexico, the Caribbean, and most of the Andes), but this change did not permeate the further reaches of the colonies, such as Central America and the Southern Cone (Cisneros E. 1998: 89; see Rona 1967; Páez Urdaneta 1981; Benavides 2003; and Weeks 2005 for a more complete history of vos). In the case of Chile, voseo dominated in educated speech until the end of the nineteenth century. During this time, Venezuelan gram- marian Andrés Bello relocated to Chile, and the use of voseo by his adopted compatriots was met by his strong disapproval. Bello and his disciples aided in securing the prominence of the tú form. The response to Bello’s criticism was quite strong; voseo was stigmatized, and tuteo was seen as the “correct” form of address (Torrejón 1986: 680). Nevertheless, the “authentic” voseo (to use Torrejón’s term, i.e., pronoun vos used with corresponding verb form) was not eradicated completely—instead the form was relegated to the lower social classes (Lipski 1994: 161). In the class that Torrejón refers to as semiculta—those who were in contact both with the lower class and the educated class—there is evidence that voseo was not completely eliminated (1986: 680). The pronoun vos was the primary target for stigmatization as it is the most salient feature of the form. This continues to be the case today (Stevenson 2007). The verb forms that correspond to voseo appear to have survived that era and are present in modern authentic voseo and mixed verbal voseo. The latter is what Lipski refers to as a crypto-voseo, in which the verb conjugation that corresponds to voseo “disguises itself” behind the subject pronoun tú (1994: 161). Thus Chilean speakers will frequently use the pronoun tú, or a null pronoun, with a verb form belonging to vos, for example, (tú) hablái instead of (tú) hablas ‘you speak’. Historically, the expansion of the mixed verbal voseo corresponds to the 1960s—a time of great social change in Chile (Benavides 2003: 620). The country experienced a simultaneous diminishing of and negative reaction against the importance of prescriptive norms (Benavides 2003: 620). In terms of the extension of verbal voseo in Chile, most studies are in agreement that the center of diffusion of the form is the capital and that it dominates from the city of Antofagasta to the south. Some studies (e.g., Benavides 2003: 620) claim that the southern island of Chiloé and the northern areas that border Bolivia and Peru prefer tuteo, whereas other investigations show variation between voseo and tuteo in these regions. The authentic voseo, in which the pronoun vos is used with corresponding verb forms, is more common in rural areas and in lower social classes in urban areas (Benavides 2003: 620).

3. Previous Studies of Chilean Voseo Voseo throughout Latin America has been widely studied (e.g., see Lapesa 1991; Rona 1967; Fontanella de Weinberg 1979; and Páez Urdaneta 1981 among many others). Most previous studies of the Chilean voseo consist of a cataloging of its varying types. In his 1967 book Geografía y morfología del voseo, Rona designates three types of voseo: type I, which maintains the original diphthongs -áis and -éis; type II, in which the original diphthongs are reduced to the strong vowels -ás and -és; and type III in which the diphthong -áis is maintained in the first conjugation but is reduced to the weak vowel-ís in the second and third conjugations. The prevailing verbal voseo in Chile corresponds most closely to Rona’s type III. According to researchers such as Torrejón (1986, 1991) and Morales Pettorino (1998), the possible forms of address for the second person singular are as follows: the authentic tuteo (pronoun tú used with the corresponding verb conjugation, such as tú tienes), the authentic voseo (pronoun vos used with the corresponding verb conjugation, such as vos tenís), the mixed pronominal voseo (pronoun vos used with a verb conjugation that corresponds with tú, such as vos tienes), and the mixed verbal voseo (pronoun tú used with a verb conjugation that corresponds with vos, such as tú tenís). Both authentic voseo and mixed verbal voseo use the same verbal inflexions, as documented in Torrejón (1986) and Morales Pettorino (1998). The present tense Chilean verbal voseo forms of regular verbs ending in -ar, -er, and -ir are llegar > llegái; comer/vivir > comís/vivís. 416 Hispania 93 September 2010

Much of the current knowledge regarding the Chilean voseo comes from Torrejón’s studies of the use of mixed verbal voseo in the educated classes in Chile. Torrejón (1986) uses his informal observations as first a student and then a professor to give an outline of the family and social relationships in which verbal voseo is given and received. He posits that the form is becoming more accepted by younger speakers while becoming less a target of stigmatization by older speakers of the upper class. It is these older speakers who made the verbal voseo popular in the educated class in the 1950s and 1960s, but according to Torrejón these speakers no longer use verbal voseo as frequently when compared to their prior use of the form. Tor- rejón attributes this surge in the popularity of verbal voseo to three major social factors: (1) a combined dissolution of the previously well-defined hierarchy with processes of social leveling, (2) young educated speakers’ desire to break with the linguistic patterns of their parents, and (3) the increased importance of institutions such as high schools and universities that served to spread the form. In his 1991 article, Torrejón further develops this outline using Brown and Gilman’s (1960) criteria of solidarity and equality. He divides Chilean society into three groups: the educated class, the working class, and the semieducated class. The semieducated class consists of an intermediate group that maintains daily contact with the groups on the extremes of the social spectrum but seeks to enter the educated class. He finds the use of the formalusted to be in line with the principles of solidarity and equality—usted is often used in situations that lack solidarity or for unequal social levels or age groups. Usted may also be used to express intimacy between romantic partners or to address children, which follows a similar pattern to other varieties of Latin American Spanish (see Lipski 1994). He also suggests that the mixed verbal voseo is common among younger speakers of the educated class—but only to a certain age, at which point they begin to prefer tuteo for informal relationships. Uber (2004) agrees, reporting that in Chile, tuteo is the most common form of address, with less frequent use of the formal usted and the mixed verbal voseo being used only in cases where there was a high level of solidarity. Torrejón concludes that the system of forms of address is becoming more simplified and egalitarian. As support for this theory, he cites the use of verbal voseo by speakers of the upper class as well as the use of tuteo and mixed verbal voseo in situations that previously called for the formal usted form. As Torrejón admits that his outline of the forms of address is based on his personal experiences and informal observations, this is among the hypotheses tested quantitatively in the current study. In a recent study, Stevenson (2007) employed role plays and matched-guise tests to determine the use and evaluation of voseo in Chile. He found that age and gender are important factors in verbal voseo and that although the pronoun vos is still widely stigmatized, its use may be increasing among young men.

4. The Study 4.1 Location and Population The sites of investigation include mostly public places in Santiago: town squares, mall food courts, and parks. The investigators were careful to include locations that would attract a population of diverse socioeconomic levels. As previous studies have pointed to the capital as the place of origin for the Chilean verbal voseo forms, all participants were native speakers residing in Santiago at the time of the study. Similar numbers of men and women participated: forty-two and thirty-eight, respectively. Their ages ranged from eighteen to seventy-one years old. The age groups used in this study are the following: young—between eighteen and thirty years old (forty-one participants), adult— between thirty-one and fifty years old (twenty-five participants), andolder —fifty-one and older (fifteen participants). The levels of education of the participants varied from incomplete primary education to postgraduate education. Bishop & Michnowicz / Forms of Address in Chilean Spanish 417

The demographic data collected—residential neighborhood, occupation, and level of education, all of which tend to correlate with income level—aided in defining the social class for each participant. The researchers used the NRS (National Readership Survey) social grade system (Miranda n.d.) that had been adapted to the specific Chilean context by considering factors such as residential neighborhood in Santiago and divided the participants into three social groups: working class, middle class, and professional class. According to the classifications, the working class (groups D and E) makes up 37% of the Chilean population. Members of the working class have not obtained a formal secondary education, and they typically work as laborers, gardeners, or domestic servants. Here, twenty-eight participants represent the working class. The middle class (group C3) forms roughly 30% of the population. Members of the middle class typically have completed secondary education and sometimes also a technical degree, which allows them to hold jobs as merchants, shopkeepers, and administrative employees. Twenty-seven participants are defined as middle class. Another about 30% of the population makes up what is denominated the professional class (groups ABC1 and C2). They usually have obtained a high level of university or postgraduate education and hold highly prestigious jobs as department managers, executives, doctors, and law- yers. Twenty-six participants represent the professional class. Thus the participant pool provides a strong cross-section of Santiago and will allow us to meet the goals of the present study.

4.2 Methodology Data were collected between December 2008 and January 2009. The primary research instrument consisted of a linguistic survey. Each participant provided his or her demographic data—age, sex, profession, education level, and residential neighborhood—and then responded to multiple-choice questions regarding with which form they would address a variety of interlocutors. The use of surveys in linguistic studies can be problematic, given the possibility that participants will answer not with the form they actually use but instead with the form they deem correct or they think the researcher wants to hear. In spite of these potential problems, a decision was made to utilize a survey for several reasons. First, it would be difficult to ob- serve all of the possible interactions with different interlocutors for each speaker in the study, although an attempt was made to do so for some interactions (see section 5.2). Thus a linguistic survey provides a point of departure for future studies using different methodologies. Second, as an initial step in the quantified study of Chilean verbal voseo, a survey allows comparison with previous studies in Chile and throughout Latin America. Finally, we argue that a survey focuses the speakers’ attention on the linguistic form being studied, and as such is an indicator of overt attitudes toward the form. In this way, surveys can bring out speakers’ evaluations of stigmatized versus unstigmatized forms. An additional attempt was made to mitigate the possible bias of a survey, as participants were encouraged to answer as if they were addressing each interlocutor and not simply to respond with the choice they deemed to be “correct.” So as long as the reader keeps the limitations of the survey in mind, the results can provide useful insight into forms of address in Chile. As previously mentioned, the authentic voseo, that is, the use of the pronoun itself, tends to be heavily stigmatized in Chile; for that reason the questionnaire in the current study includes verb conjugations corresponding to usted (formal), tú (informal), and vos (informal) instead of these subject pronouns (see Appendix). For the purposes of comparison, the choice of interlocutors for the questionnaire was based on the relationships examined in Torrejón’s investigations (1986, 1991). Two questions are devoted to each interlocutor—one with a common, high frequency verb and one with a less common, low frequency verb—in order to determine if verbal voseo is conditioned by verb frequency. A frequency dictionary of Spanish (Davies 2006) aided in classifying verb frequency. An example from the survey of a common verb is estar, and an uncommon verb tested for the 418 Hispania 93 September 2010 same interlocutor is apresurar. The survey also included two attitude questions regarding the use of verbal voseo; select responses are examined in this study. Data were coded for five external factors (interlocutor, speaker gender, age, social class, and previous residence abroad), and one linguistic factor (verb frequency). Tokens were then submitted to multivariate statistical analysis (Varbrul). The results of these analyses are detailed in section 5.1. In addition to the questionnaires, the researchers made observations and recordings of tuteo/voseo use in context during the same period. Interactions among friends, acquaintances, and strangers were observed in a variety of contexts throughout the city of Santiago. The researchers kept notes of forms of address used in situations such as client-employee relations in stores and offices, relations between co-workers, and relations between friends and family on public transportation and in family gatherings. More than 350 tokens were extracted from these notes/recordings and coded for form used, information about the situation in which the form was given, and the sex and approximate age of the interlocutors. It was hypothesized that data collected in this manner, as a corollary to the overt questionnaire data, would give a more accurate and complete picture of pronoun selection in Chilean Spanish.

5. Results 5.1. Survey Results This section details the Varbrul analyses conducted to determine what social and linguistic factors affect the form of address used for these speakers of Santiago Spanish. Table 1 presents the overall frequencies of each pronoun across factor groups. An initial review of the reported frequencies for all pronouns reveals some expected results as well as some surprising ones. First, the formal address form usted is by and large used according to pan-Hispanic norms, that is, in situations of inequality or power differential when a speaker wants to show respect or deference for her or his interlocutor (see Brown and Gilman 1960, among others). So usted dominates with grandparents (70%), professors (76%), a boss (75%), and an older employee (67%), where age is likely the deciding factor. Overall results indicate, however, that informal pronouns are generally used more frequently than formal usted, lending support to Torrejón’s (1991) observation that the system of address in Chile is becoming less hierarchical and more egalitarian. Regarding the informal pronouns, tuteo is reported far more frequently than verbal voseo, somewhat surprisingly given the authors’ impressions of Chilean Spanish. We see a general avoidance of verbal voseo with interlocutors not intimately connected to the speaker, for example, with strangers and foreigners. This mirrors the findings of previous studies of other varieties (e.g., Schreffler 1994 for El Salvador) that verbalvoseo is avoided with foreigners and nonna- tive speakers of Spanish. The contexts of use in Chile are not completely related to intimacy, however, given that significant others/spouses and co-workers receive almost the same verbal voseo, whereas children are addressed with tuteo forms. Another interesting trend observed in the data is the V-shaped distribution of vos seen across age and social class groups. These patterns are readily seen in Figure 1. Both middle-aged speakers and middle-class speakers report the lowest rates of verbal voseo, whereas young speakers and older speaker both report more verbal voseo, as do working- class and professional-class speakers. A cross-tabulation analysis comparing speaker age and social class (Figure 2) confirms these results: the two highest reported uses of verbal voseo are among older working-class speakers (19%) and young professional-class speakers (17%). When we compare these frequencies to those of middle age middle-class speakers (5% vos), we see that those two groups report more than three times more vos use than do middle groups. This V-shaped distribution suggests that verbal voseo is stigmatized among those speakers who Bishop & Michnowicz / Forms of Address in Chilean Spanish 419

Table 1. Reported Frequencies for Forms of Address

Group Factor % Usted % Tú % Vos

Interlocutor Mother 36% 51% 13% Father 31% 57% 12% Sibling 10% 59% 31% Grandparent 70% 27% 2% Boy/girlfriend – spouse 15% 64% 21% Your children 33% 60% 6% Professors 76% 24% 0% Work classmates 9% 70% 22% Children of friends 28% 65% 7% Male friend 17% 62% 20% Female friend 9% 65% 25% Stranger – same age 32% 62% 6% Foreigners (Spanish- 30% 68% 2% speaking) Foreigners (non-Spanish- 40% 60% 1% speaking) Boss 75% 25% 0% Older employee 67% 32% 1% Younger employee 32% 59% 9%

Age Young 35% 52% 13% Middle age 37% 55% 7% Older 36% 53% 10%

Gender Male 34% 53% 13% Female 38% 54% 8%

Social class Working 39% 49% 12% Middle 39% 52% 8% Professional 30% 59% 11%

Residence Lived outside of Chile 51% 42% 7% Only lived n Chile 35% 54% 11%

(Table 1 continues next page.) 420 Hispania 93 September 2010

Table 1 (continued)

Verb frequency Common 34% 54% 11% Uncommon 37% 53% 10%

Overall frequency 36% 53% 10%

Figure 1. Percentage Reported Verbal voseo for Age and Social Class have to interact daily in the workforce and who may be less secure in their social standing (see Milroy and Gordon 2003: 39). Younger and older speakers, on the one hand, and working-class and professional-class speakers, on the other, may have the social freedom to use a form that is proscribed for speakers in middle social groups. As they seek to improve their social standing by way of job stability in the workforce, speakers in the middle groups in terms of age and social class may feel more pressure to conform to the standard—in this case, tuteo—and consequently may report lower frequencies of the nonstandard form, verbal voseo. Standard forms are commonly associated with the upper class, which can lead to middle-class speakers overusing these prestige forms in an attempt to be perceived as more educated (Silva-Corvalán 2001: 106).The cross-tabulation of age and social class also reveals that the adoption of vos by professional-class speakers is a fairly recent phenomenon, as suggested by Torrejón (1986). Looking at professional-class speakers, we find an abrupt increase in reported verbalvoseo use (2% older speakers, 8% middle-age speakers, 17% younger speakers). Given that at 17%, the usage of young professional-class speakers most closely resembles that of older working-class speakers (19% vos), it appears that a form previously restricted to the working classes has made its way into the upper classes, at a rate that now exceeds the use of vos among young people of the working class (who report 11% vos use). The possibility of a change in progress or age- grading of this phenomenon will be addressed in section 6. Bishop & Michnowicz / Forms of Address in Chilean Spanish 421

Figure 2. Cross-Tabulation of Age and Social Class: Verbal Voseo

Two other cross-tabulations shed further light on the choice between tú and vos. A comparison of gender and social class reveals strong gender differences in working- and professional-class speakers but not among members of the middle class (Figure 3).

Figure 3. Cross-Tabulation of Gender and Social Class: Verbal Voseo 422 Hispania 93 September 2010

In Figure 3, we see that whereas women produce very similar frequencies of vos across social classes (9% working class, 9% middle class, 6% professional class), men vary widely, with working-class and professional-class males reporting similar frequencies for vos (15% working class, 16% professional class). This distribution further suggests that verbal voseo has moved from the working class to professionals, an extension driven, in this case, by men. Finally, previous studies on pronoun use have found important differences in frequency between interlocutors of different genders (see Schreffler 1994; Hernández 2002 among others). This is seen clearly in a cross-tabulation of gender and the interlocutors amigo and amiga

Figure 4. Cross-Tabulation of Speaker and Friend Gender: Verbal Voseo

Both genders use more vos when speaking to women than to men, although the observed difference for male speakers is not large (26% with men vs. 29% with women). The discrepancy among women speakers (14% vos use with men, 22% with women) may indicate that use of vos by a woman to a man may imply too much intimacy on the part of the speaker and could be seen as inappropriate. Thus tú is overwhelmingly preferred in these contexts (70% tú use, female speaker to a male friend). Additionally, one informant indicated that (female) speakers may at times be trying to impress the opposite sex and therefore use a form that they may perceive as “more correct,” that is, tuteo. In addition to the frequency analysis outlined above, a series of multivariate statistical analyses were conducted using GoldVarb X (Sankoff, Tagliamonte, and Smith 2005). These analyses permitted the determination of significant factors in the choice between vos and tú/ usted for these speakers of Santiago Spanish as well as the relative strength of the factors. Since the variation of interest here is the use of verbal voseo, this form was compared to the other two address forms in the multivariate analysis. Initial cross-tabulations of age and residence as well as social class and residence revealed a strong interaction. No young speakers in the corpus had lived outside of Chile; the same is true for working-class speakers. This problematic interac- Bishop & Michnowicz / Forms of Address in Chilean Spanish 423

Table 2. Multivariate Analysis Results. Application Value = Vos

Group Factor Weight Percentage Total Range

Interlocutor Sibling .85 31% 50/162 78 Female friend .81 25% 41/161 Work/classmates .77 22% 35/161 Boy/girlfriend – spouse .76 21% 34/162 Male friend .76 20% 33/128 Mother .64 13% 21/162 Father .61 12% 19/162 Younger employee .55 9% 15/161 Children of friends .49 7% 12/161 Your children .44 6% 10/161 Stranger – same age .41 6% 9/161 Grandparent .29 2% 4/162 Foreigners .18 2% 3/161 (Spanish-speaking) Foreigners (non- .13 1% 2/161 Spanish-speaking) Older employee .07 1% 1/161

Age Young .57 14% 174/1221 18 Older .50 12% 52/450 Middle-aged .39 8% 63/750

Gender Male .57 15% 184/1260 15 Female .42 9% 105/1161

Social class Working .55 14% 111/801 13 Professional .53 12% 101/810 Middle .42 9% 77/810

*Verb freq. Common [.52] 13% 156/1215 Uncommon [.48] 11% 133/1206

N = 2421; Input = .07; Log likelihood = -753.219; P = .006; X2/cell = 1.1645 424 Hispania 93 September 2010 tion with the factor group “residence outside of Chile” along with the fact that initial statistical analyses with all factor groups revealed that residence did not play a significant role in the observed variation led us to eliminate this factor from further analysis. All other factor groups were included in the analyses, the results of which are found in Table 2. In Varbrul analyses, the relative weight of each factor indicates how that factor affects the dependent variable, in this case use of verbal voseo. As a rule of thumb, factor weights greater than .5 favor the use of verbal voseo, factor weights less than .5 disfavor it, and weights of .5 are neutral. First, a comparison of factor weight range allows us to determine the relative importance of factor groups in determining the use of vos. Range is computed by subtracting the lowest factor weight from the highest factor weight within a group (see Tagliamonte 2006). With a range of 78, interlocutor is by far the most important factor in verb form selection. The analysis shows that vos use is favored with parents, siblings, significant others and spouses, co-workers or classmates, and friends. The interlocutor who most highly favors vos use is siblings (.85), followed closely by female friends (.81). On the other hand, verbal voseo was chosen less often with grandparents, children, professors, strangers, foreigners, bosses, and older employees. A younger employee shows a neutral factor weight of .55, strongly favoring none of the pronouns. Based on range, the rest of the factor groups have similar, and much less important, effects on pronoun use. Regarding age, younger speakers favor verbal voseo slightly, older speakers are neutral, and middle-age speakers disfavor it. The next factor in importance is speaker gender, with men favoring and women disfavoring verbal voseo. Finally, an analysis of social class reveals that working-class speakers use more verbal voseo, followed by professional-class, with middle-class speakers disfavoring verbal voseo. It must be remembered, however, that frequency analyses suggested that men are responsible for the rise in verbal voseo among professionals, whereas the analysis in Table 2 includes both men and women. A separate analysis with only men reveals that professional-class men favor vos (working class .57, middle class .35, professional class .58). An additional analysis with only women determined that the sole significant factor was age, with young women reporting morevoseo use (.59), whereas middle aged (.38) and older (.47) women disfavored it. Neither social class nor any of the other factors were chosen as significant by the step-up/step-down analysis for women. Finally, the one linguistic factor in the analysis, verb frequency, did not produce significant results. Again, the hypothesis was that lexical frequency plays a role in the adoption of voseo verb forms, so that a very common expression such as ‘how are you’ would be more likely expressed with voseo ¿Cómo estái? than with tuteo ¿Cómo estás?. So although the data show that verb frequency as implemented here is not a deciding factor in pronoun selection, there is a slight effect, with more frequent verbs tending to favor vos. Given the trend seen here and the copious research on the role of lexical frequency in (socio)linguistic patterning (Bybee 2002, 2007;Torres Cacoullos and Ferreira 2000 among many others), future study, perhaps with a different implementation of frequency, is needed to address this further.

5.2. Observations and Recordings Results As a corollary to the survey data outlined above, observations were made of interactions in Santiago, which were logged in a notebook. Among speakers younger than approximately age forty, verbal voseo was the most prevalent form, without noticeable difference between men and women. In the majority of the cases where the speakers used verbal voseo, it was evident that the interlocutors knew each other well, as in the case illustrated below, when two young men meet in the subway. It can be noted that much time has passed since the last time they saw each other, but they immediately employ the verbal voseo form with each other:

M1: Tanto tiempo, oye ¿estái trabajando? M2: Sí, encontré pega… (a job) Bishop & Michnowicz / Forms of Address in Chilean Spanish 425

M1: ¿Hái hablado con alguien? M2: No, ya no veo a nadie de la U(niversidad) M1: y, ¿tenís Facebook?

The majority of the interactions between young speakers who know each other well fol- low this same pattern of immediate and consistent verbal voseo, with a few isolated cases of tuteo. Among young speakers who do not know each other well, tuteo appears to be the dominant form. Some examples of this situation can be seen in stores and businesses, where the customer service context requires more formality than what is offered by verbal voseo. The following conversation took place in a department store, between a male store clerk of roughly twenty- five years of age and a thirty-year-old male client:

Store clerk: Si quieres consultar, avísame no más. Client: ¿Cuál es la diferencia entre estas dos? Store clerk: Hmm, no sé… Client: No te preocupes, total son muy caras.

A similar pattern was observed in a translation agency, but in this case the employee (female, twenty years old) first addresses the client (male, thirty years old) using tuteo and then switches to verbal voseo after the client asks her for a favor:

Client: Lo que pasa es que no sé si necesito el timbre del Ministerio de Educación. Employee: Bueno, me llamas y lo puedo agregar si lo necesitas. Client: Y cuando esté listo, ¿lo paso a buscar o…? Employee: Sí, vienes acá. Client: ¿Y para cuándo será? Lo que pasa es que me voy el 5… ¿se podría hacer más rápido? Employee: A ver… ¿para cuándo lo necesitái?

The employee switched to verbal voseo when the client requested a favor, suggesting that the employee was responding to the client’s plea with a sense of solidarity and reassurance by using the verbal voseo form. Among older speakers, verbal voseo appears in the same types of situations as it does among younger speakers, although it is less common than tuteo. In the Universidad Diego Portales, two men of roughly fifty years of age greet each other using thetú form: ¿Hola, cómo estás? A few minutes later, one makes a suggestion to the other using verbal voseo: Tú podís subirlo. The use of the formal usted seems to be more widespread in older speakers as well. Older speakers prefer usted in some situations where younger speakers would use tú, such as between a client and an employee, between two co-workers of similar age, and between two neighbors. In addition to observed uses, conversations at family gatherings were recorded in order to better understand the pattern of address. In the recorded conversations among friends and family members, a similar pattern of interactions was found. The young speakers addressed each other and their parents using verbal voseo. Often they would turn to the formal usted to address an uncle or an aunt. The older speakers addressed each other with tuteo and verbal voseo and would occasionally use usted as an affectionate form when addressing small children, perhaps as an ironic form when the speaker is clearly not showing deference to the child. The comparison of the linguistic questionnaires with the observations and recorded conversations illustrates the dramatic difference between the reported use of verbal voseo and the actual use of this form, as seen in Figure 5. In recorded/observed speech, verbal voseo accounted for 54% of the interactions, followed by tú (39%) and usted (7%). A multivariate analysis of recorded/observed tokens found that gender was not a significant factor. Age had 426 Hispania 93 September 2010 the strongest effect on use of verbal voseo, with a factor weight (f.w.) of .78 for speakers under approximately age forty, compared to an f.w. of .15 for speakers older than 40. Likewise, context was a significant factor, with more formal contexts (such as between a store employee and a customer) showing less verbal voseo (f.w. of .04), and informal contexts between friends and family evidencing more (f.w. of .53).

Figure 5. Percentage of Forms of Address: Surveys and Observations/Recorded Conver- sations

Regarding the free-response survey question on opinions of voseo, the majority mentioned some negative aspect of the form. Several participants characterized speakers who use vos as lacking in education or the ability to express themselves. Other respondents mentioned that a speaker who uses vos may be aggressive, pushy, or lacking in respect for others. Various participants responded that their opinions on the use of vos are context specific: it marks a certain intimacy in some situations but also can be used to insult or show anger. With negative attitudes such as these, it comes as no surprise that the reported frequencies of vos are much smaller than actual observed frequencies.

6. Discussion and Conclusions The data collected from the surveys and observations appear to quantitatively confirm Tor- rejón’s (1986, 1991) informal observations. It is true that the verbal voseo presents itself more strongly among young speakers and that the largest rise in frequency of use can be seen between young speakers and adult speakers, not only in family relationships as Torrejón indicates but also in global usage. A longitudinal study is needed to determine if individuals change their address forms (age-grading) or if it represents a generational shift. The present data combined with Torrejón’s (1986, 1991) studies suggest that both patterns are present in Chile. That is, Bishop & Michnowicz / Forms of Address in Chilean Spanish 427 speakers may switch to more tuteo in middle age while at the same time the global frequency of verbal voseo increases across generations. The extended use of verbal voseo as reported and observed in actual speech lends support to Torrejón’s prediction regarding the overall system of forms of address. According to Torrejón, what once was a complex system is in the process of becoming simpler and more egalitarian. As presented earlier, there is a sharp divide among older speakers: the use of verbal voseo is practically nonexistent among educated speakers (2%), whereas working-class speakers use the form with more frequency (19%). There is a much more even distribution of the form among younger speakers, and it is younger speakers of the professional class who use the form slightly more frequently than their counterparts from other social classes. As mentioned earlier, a possible stigmatization of verbal voseo may lead to lower reported frequencies on the linguistic surveys. Speakers may feel pressured to select the form they perceive as correct. The written form of the verb may be more salient to them, as the tú form is more prevalent in writing, and they may not realize the actual frequency with which they use verbal voseo. The survey data show attitudes toward forms and indicate which forms may be stigmatized. So although a comparison of survey and conversational data suggests that speakers are underreporting their use of verbal voseo, the survey provides important insight into what speakers think they do or should do. What the data here demonstrate is that although verbal voseo remains a stigmatized form in Chile, it is less stigmatized among men, young speakers, and working-class and professional-class speakers. Regarding methodology, a comparison of survey and “real-world” data suggests that linguistic surveys such as that employed here likely underreport the frequency of stigmatized forms while at the same time capturing the overall pattern of use found in recorded speech. In addition to the possible stigmatization of verbal voseo by certain groups of speakers, linguistic identity may play a role in the tendency to use the form. Previous investigations have examined the relations between certain linguistic characteristics and specific sociocultural groups, such as the postvocalic /r/ and social class in English speakers (Labov 1966) or the glottalized /s/ found among Puerto Rican adolescents who listen to reggaeton music (Valentín-Márquez 2006). According to Wolfram and Schilling-Estes (1995: 698), nonstandard varieties can be key in the creation of a cultural identity of a community. Toribio (2000) observes the role of linguistic factors in the creation of a national identity: she notes that although many Domini- cans expressed negative attitudes concerning several aspects of their own dialects, these same speakers preserved these linguistic traits even when residing outside of their native country. She posits that Dominican immigrants to the United States construct solidarity through the linguistic identity offered by their dialect. This corresponds to what Labov (1966) refers to as “covert prestige”—the use of a stigmatized nonstandard variety by a specific group to indicate solidarity or identification with that group. A similar situation can be seen in Chile. Several speakers expressed negative evaluations of the Chilean variety of Spanish, in particular the use of verbal voseo, yet they preserve these traits as part of their linguistic identity perhaps as a way to signal their group identity as Chileans. Although the actual form that verbal voseo takes in Chilean speech may be particular to Chile and therefore part of its speakers’ linguistic identity, the patterns observed in this study regarding voseo use are not exclusive to this country. For example, in his study conducted in El Salvador, Quintanilla Aguilar observes an expansion of vos use to take on functions previously reserved for tú (2009: 361). Weyers (2009) found the same pattern of vos expansion in Montevideo, Uruguay. Similarly, studies in Guatemala have shown a trilevel system of pronouns, with usted being the respectful form, vos being the intimate form, and tú being an intermediate step between the two (Pinkerton 1986: 694). Various Chilean speakers in the current study echoed opinions expressed by Guatemalans in Pinkerton’s investigation and Salvadorans in Quintanilla Aguilar’s study: tú is often seen as a more refined and “correct” form of address thanvos . Pinkerton points 428 Hispania 93 September 2010 to the predominance of vos among men as evidence for a sex-preferential phenomenon (692). This effect is not nearly as strong in Chile; however, as previously noted, men use vos slightly more than women do. Similar gender differences were also found in an analysis of several early twentieth-century theatrical scripts from the River Plate region—men tended to use vos more often than women, although, as in the present investigation, the difference between genders was not large (Moyna and Vanni Ceballos 2008: 73). An interesting parallel can also be seen in the patterns of forms of address for urban speakers in present-day Santiago and those found by Moyna and Vanni Ceballos in the turn-of-the-century plays: the young educated speakers use the vos form much more frequently than their older social counterparts do (75). As seen in the survey results and in the observations/recordings of actual speech, the Chil- ean verbal voseo appears to be a form of address used to express solidarity, especially among young speakers. What once was a highly stigmatized form on the verge of extinction in the upper and middle classes is now being promoted and widely used by young male speakers of the professional class.

ACKNOWLEDGMENTS

Author names appear in alphabetical order. The authors wish to thank the editors of His- pania and the anonymous reviewers for their prompt attention to our article and for their helpful comments. All errors remain ours.

WORKS CITED

Benavides, Carlos. (2003). “La distribución del voseo en Hispanoamérica.” Hispania 86.3: 612–23. Print. Brown, Esther L., and Rena Torres Cacoullos. (2003). “Spanish /s/: A Different Story from Beginning (Initial) to End (Final).” Amsterdam Studies in the Theory and History of Linguistic Science 4.238: 21–38. Print. Brown, Roger, and Albert Gilman. (1960). “The Pronouns of Power and Solidarity.” American Anthro- pologist 4: 24–39. Print. Bybee, Joan. (2002). “Word Frequency and Context of Use in the Lexical Diffusion of Phonetically Con- ditioned Sound Change.” Language Variation and Change 14.3: 261–90. Print. —. (2007). Frequency of Use and the Organization of Language. New York: Oxford UP. Print. Cisneros E., Mireya. (1998). “Hacia un estudio del voseo. Aspectos históricos, pragmáticos y morfológicos de los tratamientos de segunda persona singular.” Litterae 7: 76–94. Print. Davies, Mark. (2006). A Frequency Dictionary of Spanish : Core Vocabulary for Learners. New York: Routledge. Print. Fontanella de Weinberg, María Beatriz. (1979). “Algunos aspectos del voseo hispanoamericano.” Hom- enaje a Fernando Antonio Martínez: estudios de lingüística, literatura, e historia cultural. Bogota: Instituto Caro y Cuervo. 175-85. Print. Hernández, José Esteban. (2002). “Accommodation in a dialect contact situation.” Revista de Filología y Lingüística de la Universidad de Costa Rica 28.2: 93-110. Print. Labov, William. (1966). The Social Stratiﬁcation of English in New York City. Center for Applied Lin- guistics, Washington, DC. Print. Lapesa, Rafael. (1991). Historia de la lengua española. 9th ed. Madrid: Editorial Gredos. Print. Lipski, John M. (1994). Latin American Spanish. London: Longman. Print. Milroy, Lesley, and Matthew Gordon. (2003). Sociolinguistics : Method and Interpretation. Malden: Blackwell. Print. Miranda, Marcela. (n.d.). “ABCD, letras que nos dividen.” ¿Qué pasa? Web. 20 January 2010. Morales Pettorino, Félix. (1998). “La conjugación en el español de Chile.” Literatura y lingüística 11: 89–103. Web. 20 January 2010. Moyna, María Irene, and Beatriz Vanni Ceballos. (2008). “Representaciones dramáticas de una variable lingüística: tuteo y voseo en obras de teatro del Río de la Plata.” Spanish in Context 5.1: 64–88. Print. Bishop & Michnowicz / Forms of Address in Chilean Spanish 429

Páez Urdaneta, Iraset. (1981). Historia y geografía hispanoamericana del voseo. Caracas: Casa de Bello. Print. Pinkerton, Anne. (1986). “Observations on the Tú/Vos Options in Guatemalan Ladino Spanish.” Hispania 69.3: 690–98. Print. Quintanilla Aguilar, José Roberto Alexander. (2009). “Actitudes de los hablantes de San Salvador hacia el tuteo y el voseo.” Hispania 92.2: 361–73. Print. Rojas Blanco, Lillyam. (2003). “Apropos of voseo: Its History, Morphology, and Situation in Costa Rica.” Educación: Revista de la Universidad de Costa Rica 27.2: 143–63. Print. Rona, José Pedro. (1967). Geografía y morfología del voseo. Porto Alegre: Pontifícia Universidade Ca- tólica do Rio Grande do Sul. Print. Sankoff, David, Sali Tagliamonte, Eric Smith. (2005). Goldvarb X: A variable rule application for Ma- cintosh and Windows. Department of Linguistics, University of Toronto. http://individual.utoronto. ca/tagliamonte/Goldvarb/GV_index.htm. Software. Schreffler, Sandra B. (1994). “Second-Person Singular Pronoun Options in the Speech of Salvadorans in Houston, TX.” Southwest Journal of Linguistics 13.1–2: 101–19. Print. Silva-Corvalán, Carmen. (2001). Sociolingüística y pragmática del español. Washington, DC: George- town UP. Print. Stevenson, Jeffrey L. (2007). “The Sociolinguistic Variables of Chilean voseo.” Diss. U of Washington. Print. Tagliamonte, Sali A. (2006). Analysing Sociolinguistic Variation. New York: Cambridge UP. Print. Toribio, Almeida Jacqueline. (2000). “Language Variation and the Linguistic Enactment of Identity among Dominicans.” Linguistics 38.6: 1133–59. Print. Torrejón, Alfredo. (1986). “Acerca del voseo culto de Chile.” Hispania 69.3: 677–83. Print. —. (1991). “Fórmulas de tratamiento de segunda persona singular en el español de Chile.” Hispania 74.4: 1068–76. Print. Torres Cacoullos, Rena, and Fernanda Ferreira. (2000). “Lexical Frequency and Voiced Labiodental-Bilabial Variation in New Mexican Spanish.” Southwest Journal of Linguistics 19.2: 1–17. Print. Uber, Diane R. (2004). “Spanish Forms of Address in the Workplace in Latin America.” The Journal of Language for International Business 15.1: 90–99. Print. Valentín-Márquez, Wilfredo. (2006). “La oclusión glotal y la construcción lingüística de identidades so- ciales en Puerto Rico.” Selected Proceedings of the 9th Hispanic Linguistics Symposium. Ed. Nuria Sagarra and Almeida Jacqueline Toribio. Somerville: Cascadilla. 326–41. Print. Weeks, Patricia. (2005). “El voseo en Chile: factores histórico-morfológicos que explican su aparición y mantenimiento.” Diss. State U of New York-Albany. Print. Weyers, Joseph R. (2009). “The Impending Demise of Tú in Montevideo, Uruguay.” Hispania 92.4: 829–39. Print. Wolfram, Walt, and Natalie Schilling-Estes. (1995). “Moribund Dialects and the Endangerment Canon: The Case of the Ocracoke Brogue.” Language 71.4: 696–721. Print. 430 Hispania 93 September 2010

APPENDIX Questionnaire Edad: ______Sexo: ______Profesión: ______Nivel de educación: ______Lugar de nacimiento: ______Residencia en otro país: ______Por favor, indique cuál verbo usaría en las siguientes situaciones: (a) Con su madre: Se nota que estás / estái / está cansada. (b) Con su padre: ¿Me puedes / podís / puede ayudar? (c) Con su hermano/a: ¿Cómo estás / estái / está? (d) Con su abuelo/abuela: Puedes / podís / puede hacerlo. (e) Con su pololo/a o esposo/a: Eres/ erís / soi / es inteligente. (f) Con un hijo: ¿Haces / hacís / hace mucho ejercicio? (g) Con un profesor: ¿Tienes / tenís / tiene tiempo ahora? (h) Con un compañero de trabajo/ universidad: Tienes / tenís / tiene mi número de teléfono. (i) Con un hijo de un amigo: ¡Te vas / te vái / se va a caer! (j) Con un amigo: ¿Me quieres / querís / quiere acompañar? (k) Con una amiga: Siempre dices / decís / dice tonteras. (l) Con un desconocido de su edad: ¿Eres / erís / soi / es de Santiago? (m) Con un extranjero (cuya lengua es español): Sabes / sabís / sabe mucho del asunto. (n) Con un extranjero (cuya lengua no es español): ¿Sabes / sabís / sabe inglés? (o) Con su jefe: ¿Vas / vái / va a la reunión? (p) Con un empleado de mayor edad: Haces / hacís / hace un buen trabajo. (q) Con un empleado de menor edad: ¿A qué hora llegas / llegái / llega? (a) Con su madre: Cocinas / cocinái / cocina bien. (b) Con su padre: ¿Por qué me apresuras / apresurái / apresura? (c) Con su hermano/a: Ofendes / ofendís / ofende a la gente. (d) Con su abuelo/a: Almuerzas / almorzái / almuerza muy tarde. (e) Con su pololo/a o su esposo/a: Me aburres / aburrís / aburre. (f) Con un hijo: Hierves / hervís / hierve el agua. (g) Con un profesor: Entrevistas / entrevistái / entrevista a muchas personas. (h) Con un compañero de trabajo/universidad: ¿Por qué no te relajas / te relajái / se relaja? (i) Con un hijo de un amigo: ¿Trepas / trepái / trepa árboles? (j) Con un amigo: Elongas / elongái / elonga antes de hacer ejercicios. (k) Con una amiga: ¿Por qué te desesperas / te desesperái / se desespera? (l) Con un desconocido de su edad: ¿Cómo evalúas / evaluái / evalúa el Transantiago? (m) Con un extranjero (cuya lengua es español): ¿Arriendas / arrendái / arrienda un departamento? (n) Con un extranjero (cuya lengua no es español): Entrenas / entrenái / entrena por muchas horas. (o) Con su jefe: Inscribes / inscribís / inscribe gente aquí. (p) Con un/a empleado/a de mayor edad: ¿Cuándo riegas / regái / riega el jardín? (q) Con un/a empleado/a de menor edad: Revuelves / revolvís / revuelve la sopa. ¿Qué significa cuando una persona usa el voseo en vez del tuteo?